Affiliations: [a] Northeastern University, Boston, MA, USA | [b] Workday, Inc., Pleasanton, CA, USA | [c] Oracle Corporation, Redwood City, CA, USA | [d] Oracle Corporation, San Diego, CA, USA | [e] College of Engineering, Northeastern University, Boston, MA, USA
Abstract: Decision making is an important part of human activities, and is a very active area of research. In this article, a formal framework for decision making is developed. At the heart of the framework is an ontology for the decision making process that classifies both the kinds of data that are available for the decision making process and the modes of reasoning that are used to develop situation awareness and to make decisions. This ontology-driven framework resolves two outstanding problems in decision making systems: providing a theoretical foundation and declarative language support. Concerns are addressed such as providing for multiple notions of time, recording the activities used in decision making, and maintaining the provenance of both data and activities. A process model is developed that allows for flexible execution of the activities, including either synchronous or asynchronous processing and for exploring alternative or concurrent branches at any stage of the process. Detailed implementation techniques are presented. An issue tracking system is used as a running example to illustrate the concepts being developed. Use cases for the framework are given in a variety of domains including customer support, healthcare, cloud services, and the Internet of Things.
Keywords: Decision making, situation awareness, information fusion, customer service, provenance, issue tracking systems
